I took the edX Blockchain for Business class last year, and was sipping the coolaid, but now I am not as optimistic. For small groups of people and organizations, a permissioned blockchain might be over engineering when a distributed and secure shared database would be fine. For very large scale applications like health care records or smart contracts involving many thousands of individuals it makes more sense, but good luck getting off the ground.
I am still experimenting with two frameworks, one written in Rust and the other Haskell, but I am starting to think I am investing too much time.